so ive been building this direct conversion receiver for 40 meters using mostly junkbox parts and it actually receives signals now which honestly surprised me. started with a NE602 for the mixer and product detector, LM386 for audio amp, and built a VFO from a colpitts oscillator design i found in an old QST from like 2003 or something.

the problem is the RF front end is just terrible. im getting a ton of breakthrough from broadcast AM stations even though im running a 40m bandpass filter i wound myself on a T50-2 toroid. tried adjusting the coupling capacitors and i can reduce it a bit but it never totally goes away. also hearing a weird birdie around 7.150 that i think might be from the VFO getting into the mixer somehow. shielding is pretty minimal right now, just some copper tape i had laying around.

not sure if its worth trying to add a preamp stage or if that would just make the broadcast problem worse. anyone dealt with this kind of thing with the NE602? im also wondering if my bandpass filter design is just off — wound it for 7MHz center but honestly my inductor measuring setup is pretty janky so who knows if the inductance is actually right.

the 602 is notorious for its poor dynamic range, like its only got something like 15dB of IP3 which is just not great when youre sitting near strong BC stations. adding a preamp would definitely make things worse in that situation. what you probably want is actually an attenuator in front of the thing, not more gain — counterintuitive but it helps with the overload.

for the filter, if your core count and winding are off by even a turn or two you can shift the passband enough that the BC stations arent getting attenuated as much as they should be. worth measuring with whatever you have, even a basic antenna analyzer across the filter input/output can tell you where the center actually landed. also check if your filter is properly terminated — those doubly-tuned designs need to see the right source and load impedance or they fall apart.

the VFO birdie is almost certainly a shielding issue. copper tape works but the seams need to be soldered, just sticking it wont do much for RF. i had the same problem on a similar build and wrapping the VFO in its own little tin can made a huge difference.

yeah what he said about the NE602 dynamic range is real, i gave up on using it as the first RF stage on anything near a broadcast band and just use it downstream after a proper filter and sometimes a JFET preamp if the band is dead. sounds backwards but the 602 as a product detector after a proper front end works way better in my experience.

also curious what your VFO looks like physically — is it in a separate enclosure from the rest of the board or just sitting on the same ugly construction ground plane? ive found that even just a small piece of PCB material bent into a wall between the oscillator section and everything else helps a lot more than you'd think.